Ben Kizaric Student, Software Developer

As a fourth year Data Science and Computer Science student at UW Madison, I strive to learn as much as I can about these ever-changing fields.

As a full stack software developer with nearly two years cumulative full time internship experience at Northwestern Mutual, the MKE Tech Hub Coaltion, and iit-SourceTech, I continually challenge myself to exercise and improve my skills.

As a person with an interest in Entrepreneurship, I'm exited to use my skills to build projects with an impact and to solve problems big and small.

Portfolio

A project completed in partnership with the City of Madison to investigate the fairness of tax-assessed home values with respect to racial demographics.

  • Collaborated remotely with a team of students, managed deadlines, meet with stakeholders, and used version control software for datasets and data-analysis code.
  • Gained domain-specific knowledge to adequately interpret a large dataset of high-dimensional features.
  • Combined publicly available and web-scrapped datasets to examine demographic factors.
  • Performed feature engineering to create new, useful features for analysis.
  • Used a multi-variable regression model to predict the effect of many factors on correct assessment values.
  • Interpreted and performed analysis of geospatial data.
  • Generated a wide array of intermediary and final visualizations using matplotlib.
  • Presented work to stakeholders, and created a written report.

Tools & Technologies Used:

Python, PandasMatplotlibFeature EngineeringRegressionClusteringVirtual-Environments
PHOSVD Image Compression March 2020 - Ongoing

A new image compression format that uses both unsupervised and supervised machine learning techniques to (hopefully) outperform existing methods of image compression

  • Utilized principle component analysis to optimize color information in sub-blocks of image.
  • Developed novel unsupervised learning technique for learning multiple orthogonal sub-spaces of high-dimensional data using gaussian mixture models and expectation maximization.
  • Used cross-validation to tune the hyper-parameters of models used for different stages of image compression.
  • Performed analysis, comparing performance to existing image compression standards like JPEG and PNG.
  • Leveraged numpy and vectorization techniques to improve runtime performance of training compression models.

Tools & Technologies Used:

PythonNumpysklearnMatplotlibCross-ValidationVirtual EnvironmentsKMeans ClusteringGaussian Mixture ModelsJupyter
MKE Tech Pulse June 2020 - Ongoing

A showcase of metrics tracking Milwaukee's status as a Tech Hub

  • Worked directly with stakeholders from multiple organizations to gather, analyze, and display data on a custom-made public portal, managed by a proprietary internal website
  • Organized meetings, managed ongoing and future tasks, and mentored members of a six-person interdisiplinary team
  • Conducted principal development work on the aforementioned web software, including creating a Vue-TS front-end, TS-Node-Express back-end and SQL Server database
  • Setup continuous deployment using AWS CodePipeline, AWS Elastic Beanstalk, and AWS S3
  • Created over 10 interactive data visualizations using Microsoft Power BI

Tools & Technologies Used:

AWS CodePipelineAWS Elastic BeanstalkAWS S3NodeJSJavascriptTypescriptVueJSjQueryExpressSQLSQL ServerSelenuim

My final project for an introductory Data Science class, an analysis of Horror Movie data

  • Gathered data from a selenuim-built web scraper.
  • Used popular python data analysis and visualization tools
  • Created ten visualizations on different aspects of the data, with a focus on analysing the relationship between movie ratings, popularity, and profitability.
  • Published to a static website.

Tools & Technologies Used:

PythonMatplotlibPandasSciKitLearnSelenuim
Swenden May 2019

A full-stack NodeJS web app with basic social media functionality

  • Implemented a front end with VueJS.
  • Implemented a back end with sign-in, sign-up, and content creation in NodeJS with typescript and express.
  • Built and deployed the app to Heroku.
  • Hosted a SQL Server database in Microsoft Azure.
  • Collected user feedback

Tools & Technologies Used:

NodeJSJavascriptTypescriptVueJSExpressSQLSQL ServerAzureHeroku
Flexcards (Prototype) November 2018 - March 2019

A Prototype of an advanced study tool for the 2019 innovation competition Transcend Madison

  • Won Best Pitch (Early Stage).
  • Developed a new system for memorization that enables more granular recall.
  • Created a desktop app as a minimal viable product.
  • Validated the software in a partnership with a UW professor.
  • Piched the idea to industry veterans.

Tools & Technologies Used:

NodeJSElectronTypescript
Lazy Math May 2018

My final project for AP Calculus and AP CS Principles. For computation & visualization of Derivatives and Slope Feilds

  • Used a custom expression tree for computation, no math libraries used.
  • Expressed slope fields as a heatmap.
  • Solved for the derivatives of 1-variable functions.

Tools & Technologies Used:

ElectronJavascriptjQuery